Comparison of Microstructures of Magnetite Reduced by H 2 and CO under Microwave Field

: The reduction of magnetite in H 2 and CO atmospheres was compared using a microwave-heating technique. The reduction of magnetite in a mixed H 2 + CO atmosphere was compared with respect to the effects of a microwave ﬁeld and a conventional ﬁeld. Microstructural changes were observed using an electron microscope. The results show that the metallization rate and reduction degree of the H 2 -reduced magnetite are much higher than those of the magnetite reduced by CO at 900–1100 °C. The Fe phase generated by H 2 reduction forms a cavity structure, and the Fe phase generated by CO reduction forms a dense block. Under conventional heating conditions, the higher the H 2 content in a pure CO atmosphere, the better the reduction effect. Under the effect of a microwave ﬁeld, the atmosphere with the highest reduction rate was 50% H 2 + 50% CO. Compared with conventional heating, the bubble holes formed by reduced iron in microwave ﬁeld are larger under the same conditions.


Introduction
The steel industry is facing increasingly severe environmental and policy pressures regarding carbon emissions. The world's leading steel manufacturers have incorporated carbon reduction into their national development strategies since the beginning of the 21st century. The existing approach to iron production is essentially a "carbon metallurgy" process using a carbon base as a reducing agent and a heat source. Exploring more innovative methods and breakthrough technologies is of great significance with respect to achieving ultra-low carbon emissions in the ironmaking stage. The rational use of H 2 in the process of ironmaking has a more significant effect on reducing energy consumption and CO 2 emissions. People continue to carry out experimental research on and theoretical simulations of various iron ore products (pellets, sinter, ore powder) under various atmospheres [1][2][3][4][5][6]. It is generally believed that the reaction rate and diffusion capacity of iron ore or pellets under a CO reduction atmosphere are much lower than those under the conditions relating to H 2 [7][8][9][10][11]. In cases of the same reduction degree, the direct reduced iron (DRI) yield of H 2 is lower than that of CO. The existing direct reduction process has a low production capacity, and its reduction temperature cannot be higher than 850°C (with this limit being imposed by sintering at high temperature).
Microwave processing has the characteristics of high-efficiency electric-thermal conversion, selective heating, volumetric heating, and an inverse thermal gradient. Furthermore, many experimental results have shown that microwaves enhance diffusion and reaction [12,13]. The "electric-hydrogen combination" model, which uses hydrogen instead of carbon for efficient reduction and electricity instead of carbon for a heat supply, is one method for eliminating carbon emission pressure. Microwave treatment enhances the permeability of a reaction gas and circumvents the premature closure of surface pores under a conventional thermal field [14]. Microwave radiation can change the thermodynamic equilibrium [15,16], and it has a significant impact on the rate of chemical reaction and diffusion [17,18]. At present, the microwave heating reduction of iron ore has become a focus of solid carbothermal reduction research. Stir [19] studied the reduction of magnetite by carbon black under microwave radiation at 1050-1100 • C using online time-resolved XRD. The reduction of Fe 3 O 4 is governed by a phase boundary reaction. The dynamic aggregation of cation vacancies and structural defects affects the kinetics of FeO's reduction to porous Fe. Amini A [20] studied the effect of the ratio of magnetite particle size to microwave penetration depth on the H 2 reduction reaction behavior. The results showed that the reduction effect of large particles (>40 µm) is better than that of small particles. However, there are few reports on the micro-morphology of the H 2 /CO reduction of magnetite under a microwave field. This manuscript compares the reduction effect and microstructures of magnetite processed using microwave-heating technology under different conditions in H 2 and CO atmospheres. The reduction effect and microstructural differences of magnetite in a mixed H 2 + CO atmosphere influenced by a microwave field and a conventional field were compared.

Experimental Equipment
The microwave-heating instrument used in this experiment was a WBMW-G4 tubular microwave-heating furnace (Tangshan Renshi Juyuan Microwave Instrument, Tangshan, China). The main components used are as follows: a heating furnace, a controllable atmosphere cabinet, a mixing box, and a high-temperature-resistant quartz tube. The temperature limit of the high-temperature-resistant quartz tube was ≤1500 • C, and its dimensions were Φ30 × 1000 mm. The heating constant temperature zone was 300 mm. The frequency of this furnace is 2.45 GHz, its power level is 4 kW, and the heating rate applied was 20°C/min. The experimental samples were placed in a corundum crucible (60 mm × 30 mm × 15 mm) without compaction. As shown in Figure 1., the gas flow rate control instrument was placed in front of the experimental furnace, allowing for the more accurate control of gas flow. The gasses used in the experiment were Ar, H 2 , and CO with a purity of 99.9% provided by Tangshan Rong sheng Gas (Tangshan Renshi Juyuan Microwave Instrument, Tangshan, China). The total gas flow rate was 1660 mL/min, and the reducing gas was mixed with a protective gas (Ar) in different proportions. Argon gas from the same supplier with the same purity was used for purging before and after reduction. The conventional heating furnace used in the experiment has been employed in previous studies [21]. Magnetite powder was embedded in JZXM0232 conductive hot inlay (conductive material was a copper-based polymer). The embedding temperature was 140 • C, the holding time was 10 min, and the mold was removed after it had cooled to room temperature. Then, grinding and polishing were performed to observe the cross sections of the mineral powder particles. Scanning electron microscopy (SEM, TESCAN MIRA LMS, Tangshan, China) was used to collect microstructural observations. a conventional thermal field [14]. Microwave radiation can change the thermodynamic equilibrium [15,16], and it has a significant impact on the rate of chemical reaction and diffusion [17,18]. At present, the microwave heating reduction of iron ore has become a focus of solid carbothermal reduction research. Stir [19] studied the reduction of magnetite by carbon black under microwave radiation at 1050-1100 °C using online time-resolved XRD. The reduction of Fe3O4 is governed by a phase boundary reaction. The dynamic aggregation of cation vacancies and structural defects affects the kinetics of FeO's reduction to porous Fe. Amini A [20] studied the effect of the ratio of magnetite particle size to microwave penetration depth on the H2 reduction reaction behavior. The results showed that the reduction effect of large particles (>40 μm) is better than that of small particles. However, there are few reports on the micro-morphology of the H2/CO reduction of magnetite under a microwave field. This manuscript compares the reduction effect and microstructures of magnetite processed using microwave-heating technology under different conditions in H2 and CO atmospheres. The reduction effect and microstructural differences of magnetite in a mixed H2 + CO atmosphere influenced by a microwave field and a conventional field were compared.

Experimental Equipment
The microwave-heating instrument used in this experiment was a WBMW-G4 tubular microwave-heating furnace (Tangshan Renshi Juyuan Microwave Instrument, Tangshan, China). The main components used are as follows: a heating furnace, a controllable atmosphere cabinet, a mixing box, and a high-temperature-resistant quartz tube. The temperature limit of the high-temperature-resistant quartz tube was ≤1500 °C, and its dimensions were Φ30 × 1000 mm. The heating constant temperature zone was 300 mm. The frequency of this furnace is 2.45 GHz, its power level is 4 kW, and the heating rate applied was 20 ℃/min. The experimental samples were placed in a corundum crucible (60 mm × 30 mm × 15 mm) without compaction. As shown in Figure 1., the gas flow rate control instrument was placed in front of the experimental furnace, allowing for the more accurate control of gas flow. The gasses used in the experiment were Ar, H2, and CO with a purity of 99.9% provided by Tangshan Rong sheng Gas (Tangshan Renshi Juyuan Microwave Instrument, Tangshan, China). The total gas flow rate was 1660 mL/min, and the reducing gas was mixed with a protective gas (Ar) in different proportions. Argon gas from the same supplier with the same purity was used for purging before and after reduction. The conventional heating furnace used in the experiment has been employed in previous studies [21]. Magnetite powder was embedded in JZXM0232 conductive hot inlay (conductive material was a copper-based polymer). The embedding temperature was 140 °C, the holding time was 10 min, and the mold was removed after it had cooled to room temperature. Then, grinding and polishing were performed to observe the cross sections of the mineral powder particles. Scanning electron microscopy (SEM, TESCAN MIRA LMS, Tangshan, China) was used to collect microstructural observations.

Experimental Raw Materials
Magnet mineral powder was used as a raw material in the experiment. The magnetite used in the experiment was acquired from a steel plant. The weight of each experimental sample was 15 g. An electronic balance (Changshu Shuang jie Testing Instrument Factory JJ224BC type, Changshu, China) was used to accurately weigh the samples. The samples were dried before the experiment. The drying temperature was 180 • C, and the drying time was 2 h. An X-ray diffractometer (Rigaku SmartLab SE, Tokyo, Japan) was used to quantitatively detect the magnetite concentrate used in the experiment. A table presenting the composition of the mineral powder is shown in Table 1. The magnet powder was dried at 180 • C for 4 hours using a drying instrument. The samples were tested using a Mastersizer 2000 laser particle size analyzer (Malvern, Shanghai, China) ( Figure 2). The particle size of the magnet powder used in this experiment was between 10 µm and 200 µm, and the median particle size (D50) was 60.256 µm.

Experimental Raw Materials
Magnet mineral powder was used as a raw material in the experiment. The magnetite used in the experiment was acquired from a steel plant. The weight of each experimental sample was 15 g. An electronic balance (Changshu Shuang jie Testing Instrument Factory JJ224BC type, Changshu, China) was used to accurately weigh the samples. The samples were dried before the experiment. The drying temperature was 180 °C, and the drying time was 2 h. An X-ray diffractometer (Rigaku SmartLab SE, Tokyo, Japan) was used to quantitatively detect the magnetite concentrate used in the experiment. A table presenting the composition of the mineral powder is shown in Table 1. The magnet powder was dried at 180 °C for 4 hours using a drying instrument. The samples were tested using a Mastersizer 2000 laser particle size analyzer (Malvern, Shanghai, China) ( Figure 2). The particle size of the magnet powder used in this experiment was between 10 μm and 200 μm, and the median particle size (D50) was 60.256 μm.

Calculation of Metallization Rate and Reduction Degree
Chemical titration was used to measure the metallic iron (MFe) and total iron (TFe) content, and then the metallization rate was calculated (Formula (1)): where m is the degree of metallization, and TFe and MFe are the total iron and metallic iron fractions in the powder sample after reduction time t, respectively. The weight loss measurement method was used to calculate weight loss, and then the reduction degree was obtained. Before the experiment, the undried magnetite sample was weighed and denoted as m1. Then, the samples were dried and dehydrated. After drying, the samples were weighed and denoted as m2. The reduced magnetite at the end

Calculation of Metallization Rate and Reduction Degree
Chemical titration was used to measure the metallic iron (MFe) and total iron (TFe) content, and then the metallization rate was calculated (Formula (1)): where m is the degree of metallization, and TFe and MFe are the total iron and metallic iron fractions in the powder sample after reduction time t, respectively. The weight loss measurement method was used to calculate weight loss, and then the reduction degree was obtained. Before the experiment, the undried magnetite sample was weighed and denoted as m 1 . Then, the samples were dried and dehydrated. After drying, the samples were weighed and denoted as m 2 . The reduced magnetite at the end of the experiment was weighed again and denoted as m 3 . m 1 − m 2 , denoted as m 4 , represent the mass of lost water. m 4 − m 3 signifies the mass of oxygen lost in the reduction process. The degree of reduction was calculated as follows: Figure 3a shows a comparison of the metallization rates of magnetite reduced by a microwave field. The reducing atmosphere employed consisted of 30% H 2 -Ar, 30% CO-Ar, 60% H 2 -Ar, and 60% CO-Ar. When the reduction atmosphere concentration was 30% and the temperature was 900-1100°C, the metallization rate of H 2 reduction was 30-50% higher than that of CO reduction, and the maximum difference was 50% at 1050°C. When the reduction atmosphere concentration was 60% and the temperature was 900-1100°C, the metallization rate of H 2 reduction was 43-56% higher than that of CO reduction, and the maximum difference was 56% at 1100°C. The higher the concentration of the reducing atmosphere and the higher the temperature, the more significant the metallization rate of the H 2 -reduced magnetite compared to that of the CO-reduced magnetite. The metallization rate of the magnetite reduced for 40 min by 60% H 2 -Ar at 1100 • C was 94%, and the reduction degree was close to 1. Under the same conditions, the metallization rate of CO reduction was 38%, and the reduction degree was 0.51. microwave field. The reducing atmosphere employed consisted of 30% H2-Ar, 30% CO-Ar, 60% H2-Ar, and 60% CO-Ar. When the reduction atmosphere concentration was 30% and the temperature was 900-1100 ℃, the metallization rate of H2 reduction was 30-50% higher than that of CO reduction, and the maximum difference was 50% at 1050 ℃. When the reduction atmosphere concentration was 60% and the temperature was 900-1100 ℃, the metallization rate of H2 reduction was 43-56% higher than that of CO reduction, and the maximum difference was 56% at 1100 ℃. The higher the concentration of the reducing atmosphere and the higher the temperature, the more significant the metallization rate of the H2-reduced magnetite compared to that of the CO-reduced magnetite. The metallization rate of the magnetite reduced for 40 min by 60% H2-Ar at 1100 °C was 94%, and the reduction degree was close to 1. Under the same conditions, the metallization rate of CO reduction was 38%, and the reduction degree was 0.51.

Reduction Effect Analysis
The effect of different H2/CO concentrations on the degree of reduction of reduced magnetite under microwave heating between 900 and 1100 °C is shown in Figure 3b. At 1100 °C, the degree reduction of magnetite reaches 1, reaching 0.65 in 40 min with 60% and 30% H2. At 1100 °C, the degree of reduction of magnetite reaches 0.51 and 0.32 in 40 min when reduced by 60% and 30% CO, respectively. With an increasing CO concentration, the degree of reduction of magnetite after reduction was significantly improved.
Similar conclusions have been obtained in studies of conventional fields. The rate at which solid iron oxide is reduced by H2 can reach 5 to 10 times that of the CO reduction rate, and this effect is more significant when the temperature increases [22]. At 500 ℃, the mutual diffusion coefficient of H2/H2O gas in iron-ore-sintered particles is 3 times that of CO/CO2 [23]. In the molten state, the rate at which FeO is reduced by H2 is 20 times that of the solid state [24]. Hydrogen molecules are much smaller than carbon monoxide molecules and can penetrate more deeply into the crystal structure of iron oxide, resulting in a greater degree of metallization. Both high temperature and increasing H2% in the atmosphere help to improve the range of interfacial chemical reaction control and reduce the range of diffusion influence in the reduction process [6].   The effect of different H 2 /CO concentrations on the degree of reduction of reduced magnetite under microwave heating between 900 and 1100 • C is shown in Figure 3b. At 1100 • C, the degree reduction of magnetite reaches 1, reaching 0.65 in 40 min with 60% and 30% H 2 . At 1100 • C, the degree of reduction of magnetite reaches 0.51 and 0.32 in 40 min when reduced by 60% and 30% CO, respectively. With an increasing CO concentration, the degree of reduction of magnetite after reduction was significantly improved.
Similar conclusions have been obtained in studies of conventional fields. The rate at which solid iron oxide is reduced by H 2 can reach 5 to 10 times that of the CO reduction rate, and this effect is more significant when the temperature increases [22]. At 500°C, the mutual diffusion coefficient of H 2 /H 2 O gas in iron-ore-sintered particles is 3 times that of CO/CO 2 [23]. In the molten state, the rate at which FeO is reduced by H 2 is 20 times that of the solid state [24]. Hydrogen molecules are much smaller than carbon monoxide molecules and can penetrate more deeply into the crystal structure of iron oxide, resulting in a greater degree of metallization. Both high temperature and increasing H 2 % in the atmosphere help to improve the range of interfacial chemical reaction control and reduce the range of diffusion influence in the reduction process [6].

Magnetite Particle Morphology
The magnetite samples and epoxy resins were mounted with a mounting instrument and then polished. The morphology of the particles was observed using a scanning electron microscope (SEM). Figure 4a is an SEM image of the cross section of the magnetite particles. The gray regions correspond to Fe 3 O 4 , and the bright white regions correspond to the Cu in the inlay. Figure 4b depicts the original particle morphology.

Magnetite Particle Morphology
The magnetite samples and epoxy resins were mounted with a mounting instrument and then polished. The morphology of the particles was observed using a scanning electron microscope (SEM). Figure 4a is an SEM image of the cross section of the magnetite particles. The gray regions correspond to Fe3O4, and the bright white regions correspond to the Cu in the inlay. Figure 4b depicts the original particle morphology.  Figure 5 presents an image of the cross section of the magnetite particles reduced by 30% H2-Ar and influenced by a microwave field. The reduction time was 40 min. In the figure, FeO is encircled by a reduced Fe phase. The metal iron layer that was reduced at a low temperature is dense. With the increase in temperature, the degree of reduction increases, the porous structure of metallic iron is obtained, and the iron layer is thickened. The reduction rate determines the formation of pores. The reaction rate increases with the increase in temperature. H2 is more likely to penetrate the iron shell into FeOx. The result of gas permeation leads to the generation of bubbles, and the generation of bubbles further leads to tearing. The formation of water and its direct evaporation to the shell may be one of the reasons for the formation of pores.  Figure 6 shows the image of 30% CO-Ar reduction magnetite in microwave field. The reduced metal iron layer is dense and blocky around the particles. The reduced metal iron layer is dense and non-porous. The metallic iron in the carbon monoxide reduction products is distributed in large aggregates in mineral particles. Figures 5 and 6 present the reduction degrees of the magnetite reduction products reduced by H2 and CO at different temperatures, respectively. When comparing the microstructure of the H2 reduction products in Figures 5 and 6, it can be seen that the reduction rate of CO is significantly slower than that of H2 reduction. Under the same conditions, H2 reduction produced more Fe than CO reduction. The unreduced FeO was completely surrounded by the generated  Figure 5 presents an image of the cross section of the magnetite particles reduced by 30% H 2 -Ar and influenced by a microwave field. The reduction time was 40 min. In the figure, FeO is encircled by a reduced Fe phase. The metal iron layer that was reduced at a low temperature is dense. With the increase in temperature, the degree of reduction increases, the porous structure of metallic iron is obtained, and the iron layer is thickened. The reduction rate determines the formation of pores. The reaction rate increases with the increase in temperature. H 2 is more likely to penetrate the iron shell into FeOx. The result of gas permeation leads to the generation of bubbles, and the generation of bubbles further leads to tearing. The formation of water and its direct evaporation to the shell may be one of the reasons for the formation of pores.

Magnetite Particle Morphology
The magnetite samples and epoxy resins were mounted with a mounting instrument and then polished. The morphology of the particles was observed using a scanning electron microscope (SEM). Figure 4a is an SEM image of the cross section of the magnetite particles. The gray regions correspond to Fe3O4, and the bright white regions correspond to the Cu in the inlay. Figure 4b depicts the original particle morphology.  Figure 5 presents an image of the cross section of the magnetite particles reduced by 30% H2-Ar and influenced by a microwave field. The reduction time was 40 min. In the figure, FeO is encircled by a reduced Fe phase. The metal iron layer that was reduced at a low temperature is dense. With the increase in temperature, the degree of reduction increases, the porous structure of metallic iron is obtained, and the iron layer is thickened. The reduction rate determines the formation of pores. The reaction rate increases with the increase in temperature. H2 is more likely to penetrate the iron shell into FeOx. The result of gas permeation leads to the generation of bubbles, and the generation of bubbles further leads to tearing. The formation of water and its direct evaporation to the shell may be one of the reasons for the formation of pores.  Figure 6 shows the image of 30% CO-Ar reduction magnetite in microwave field. The reduced metal iron layer is dense and blocky around the particles. The reduced metal iron layer is dense and non-porous. The metallic iron in the carbon monoxide reduction products is distributed in large aggregates in mineral particles. Figures 5 and 6 present the reduction degrees of the magnetite reduction products reduced by H2 and CO at different temperatures, respectively. When comparing the microstructure of the H2 reduction products in Figures 5 and 6, it can be seen that the reduction rate of CO is significantly slower than that of H2 reduction. Under the same conditions, H2 reduction produced more Fe than CO reduction. The unreduced FeO was completely surrounded by the generated  Figure 6 shows the image of 30% CO-Ar reduction magnetite in microwave field. The reduced metal iron layer is dense and blocky around the particles. The reduced metal iron layer is dense and non-porous. The metallic iron in the carbon monoxide reduction products is distributed in large aggregates in mineral particles. Figures 5 and 6 present the reduction degrees of the magnetite reduction products reduced by H 2 and CO at different temperatures, respectively. When comparing the microstructure of the H 2 reduction products in Figures 5 and 6, it can be seen that the reduction rate of CO is significantly slower than that of H 2 reduction. Under the same conditions, H 2 reduction produced more Fe than CO reduction. The unreduced FeO was completely surrounded by the generated Fe phase, and the outer Fe phase was reduced to porous iron layer by layer. CO was reduced at the defect site and gradually expanded to form dense iron. There is no complete coating phase in the CO reduction product. S.P. Matthew [25] studied the surface morphology changes in the reduction process of dense Wüstite using an SEM detection method. Instabilities develop on the oxide surface prior to the growth of a dense iron layer. These instabilities develop into pores that continue to advance into the bulk oxide before advancing into the iron layer. During the reduction process, the formation of porous iron products at the interface mainly depends on the gas composition, reaction temperature, and mineral composition. It can be seen from Figure 5 that under a 30% H 2 atmosphere, the increase in temperature is beneficial to the formation of a porous structure, and the coating layer forms more easily at a low temperature (900 • C), while the small particle ore powder has almost no pores. In the reduction of ore powder by CO, the change in temperature and CO partial pressure could not produce either coated layered iron or pore structures. The above phenomenon shows that the formation of porous structures is affected by temperature and partial pressure. In addition, the effect of gas phase diffusion as a limiting factor of reduction is very critical. It is not easy to produce a porous structure when the gas phase diffusion is limited. In the process of H 2 reduction, hydrogen migrates to the interface, and an interface reaction occurs after the migration to the interface. The interface gradually shrinks with the extension of the reduction time. Since hydrogen has no orientation dependence, it advances as a whole at the reduction interface. CO reduction is reduced at the defect site, and the adsorption of CO responds accordingly. As the reduction proceeds, iron diffuses at the defect site, thereby forming a dense iron block. In the process of CO reduction, a carbon evolution reaction was more likely to occur at a low temperature in the temperature range of this experiment [26]. Therefore, the carbon content produced at 900 • C was relatively high. The resulting reduction product has high hardness and is easy to break. The reduction product is obviously broken at 900 • C. Fe phase, and the outer Fe phase was reduced to porous iron layer by layer. CO was reduced at the defect site and gradually expanded to form dense iron. There is no complete coating phase in the CO reduction product. S.P. Matthew [25] studied the surface morphology changes in the reduction process of dense Wüstite using an SEM detection method. Instabilities develop on the oxide surface prior to the growth of a dense iron layer. These instabilities develop into pores that continue to advance into the bulk oxide before advancing into the iron layer. During the reduction process, the formation of porous iron products at the interface mainly depends on the gas composition, reaction temperature, and mineral composition. It can be seen from Figure 5 that under a 30% H2 atmosphere, the increase in temperature is beneficial to the formation of a porous structure, and the coating layer forms more easily at a low temperature (900 °C), while the small particle ore powder has almost no pores. In the reduction of ore powder by CO, the change in temperature and CO partial pressure could not produce either coated layered iron or pore structures. The above phenomenon shows that the formation of porous structures is affected by temperature and partial pressure. In addition, the effect of gas phase diffusion as a limiting factor of reduction is very critical. It is not easy to produce a porous structure when the gas phase diffusion is limited. In the process of H2 reduction, hydrogen migrates to the interface, and an interface reaction occurs after the migration to the interface. The interface gradually shrinks with the extension of the reduction time. Since hydrogen has no orientation dependence, it advances as a whole at the reduction interface. CO reduction is reduced at the defect site, and the adsorption of CO responds accordingly. As the reduction proceeds, iron diffuses at the defect site, thereby forming a dense iron block. In the process of CO reduction, a carbon evolution reaction was more likely to occur at a low temperature in the temperature range of this experiment [26]. Therefore, the carbon content produced at 900 °C was relatively high. The resulting reduction product has high hardness and is easy to break. The reduction product is obviously broken at 900 °C. In addition, the appearance of pores on the particle surface is conducive to the diffusion of the gas generated by the reaction to the interface for the reaction and the expulsion of the product gas. Zhan D [27] investigated the reduction of fine iron ore in a pure H2 atmosphere. The angle between the surface and the Fe/Fe1-x interface is smaller than the angle reduced in the pure CO. Therefore, the migration paths of V // Fe and 2e --in pure H2 are shorter than those in pure CO, which causes the Fe/Fe1-xO interface to move faster in pure H2. The results obtained by Yamashita [28] are consistent with this phenomenon, proving that when Fe0.94O is reduced at different temperatures and different reduction potentials, if the angle between the surface and the reaction interface is small, the migration speed of V // Fe and 2e --will increase, and the moving speed of the Fe/Fe0.94O interface will also increase. As shown in Figure 7, the microstructure of the particles after reduction at 1000 °C for 40 min. The mineral particles after H2 reduction are porous honeycomb. The interconnected pore structure allows the gas to continuously enter the reaction interface from the sample surface through the product layer. However, there are many iron whiskers on the surface of the mineral after CO reduction, and the whiskers and dense layer are not conducive to efficient reduction. In addition, the appearance of pores on the particle surface is conducive to the diffusion of the gas generated by the reaction to the interface for the reaction and the expulsion of the product gas. Zhan D [27] investigated the reduction of fine iron ore in a pure H 2 atmosphere. The angle between the surface and the Fe/Fe 1-x interface is smaller than the angle reduced in the pure CO. Therefore, the migration paths of V // Fe and 2e − in pure H 2 are shorter than those in pure CO, which causes the Fe/Fe 1-x O interface to move faster in pure H 2 . The results obtained by Yamashita [28] are consistent with this phenomenon, proving that when Fe 0.94 O is reduced at different temperatures and different reduction potentials, if the angle between the surface and the reaction interface is small, the migration speed of V // Fe and 2e − will increase, and the moving speed of the Fe/Fe 0.94 O interface will also increase. As shown in Figure 7, the microstructure of the particles after reduction at 1000 • C for 40 min. The mineral particles after H 2 reduction are porous honeycomb. The interconnected pore structure allows the gas to continuously enter the reaction interface from the sample surface through the product layer. However, there are many iron whiskers on the surface of the mineral after CO reduction, and the whiskers and dense layer are not conducive to efficient reduction.

Reduction Effect Analysis
For the analysis of microwave heating and conventional heating at 1000 ℃ under the conditions of a H2 + CO gas mixture employed for the reduction of magnetite, the following reducing atmospheres were used: 100% CO, 30% H2 + 70% CO, 40%H2 + 60% CO, 50% H2 + 50% CO, 60% H2 + 40% CO, 70% H2 + 30% CO, and 100% H2. Figure 8a shows that under conventional heating experimental conditions, the atmosphere with the lowest reduction degree was 100% CO. As long as H2 was added in a pure CO atmosphere, the reduction effect was higher than that in a pure CO atmosphere, and the higher the H2 content, the better the reduction effect. The reason for this involves the deposition of carbon. The presence of a small amount of CO can provide energy to the H2-reduced iron ore, making the internal temperature of the mineral particles slightly higher than that of pellets without CO, which is conducive to reducing the interfacial chemical resistance of H2 reduction. Geassy [29] investigated the reduction of Fe2O3 compacts using a H2 + CO mixture in the temperature range of 700-1100 °C. Adding H2 to CO can increase the reduction rate. The reduction rate of pure H2 gas is faster than that of pure CO. Moon [30] investigated the reduction kinetics of hematite compacted at 800 °C and 950 °C with H2 + CO mixtures using a thermogravimetric analyzer. The reaction rate increased with the increase in the H2 content in the mixture. Olsson RG [31] showed that low concentrations of hydrogen also greatly increased the decomposition rate of CO, which might have been due to the catalytic effect of adsorbing hydrogen on iron. At high concentrations of hydrogen, the carbon deposition rate slows down due to the following reverse reaction: H2O + C→H2 + CO.
Under a microwave field, it is not the case that the more H2 added to the CO atmosphere, the better the reduction effect. The atmosphere with the highest reduction was 50% H2 + 50% CO (Figure 8b). The reason for this finding is most likely that CO changes the partial pressure of the atmosphere. CO and H2O are polar molecules, and polar molecules are susceptible to microwave radiation, which changes the balance of CO + H2O→H2 + CO2, making the reduction reaction continue.

Reduction Effect Analysis
For the analysis of microwave heating and conventional heating at 1000°C under the conditions of a H 2 + CO gas mixture employed for the reduction of magnetite, the following reducing atmospheres were used: 100% CO, 30% H 2 + 70% CO, 40%H 2 + 60% CO, 50% H 2 + 50% CO, 60% H 2 + 40% CO, 70% H 2 + 30% CO, and 100% H 2 . Figure 8a shows that under conventional heating experimental conditions, the atmosphere with the lowest reduction degree was 100% CO. As long as H 2 was added in a pure CO atmosphere, the reduction effect was higher than that in a pure CO atmosphere, and the higher the H 2 content, the better the reduction effect. The reason for this involves the deposition of carbon. The presence of a small amount of CO can provide energy to the H 2 -reduced iron ore, making the internal temperature of the mineral particles slightly higher than that of pellets without CO, which is conducive to reducing the interfacial chemical resistance of H 2 reduction. Geassy [29] investigated the reduction of Fe 2 O 3 compacts using a H 2 + CO mixture in the temperature range of 700-1100 • C. Adding H 2 to CO can increase the reduction rate. The reduction rate of pure H 2 gas is faster than that of pure CO. Moon [30] investigated the reduction kinetics of hematite compacted at 800 • C and 950 • C with H 2 + CO mixtures using a thermogravimetric analyzer. The reaction rate increased with the increase in the H 2 content in the mixture. Olsson RG [31] showed that low concentrations of hydrogen also greatly increased the decomposition rate of CO, which might have been due to the catalytic effect of adsorbing hydrogen on iron. At high concentrations of hydrogen, the carbon deposition rate slows down due to the following reverse reaction: H 2 O + C→H 2 + CO.  Figures 9 and 10 show microstructural images of magnetite reduced by a mixed H2 + CO atmosphere under a microwave field and conventional field, respectively. The edge of the Fe layer reduced by pure H2 under a microwave field is thicker, and the Fe layer reduced by 60% H2 + 40% CO is highly dispersed. Even if the particles are relatively small, the reduced Fe is highly dispersed. That is, the pores generated by the reduction of mineral powder particles with a high reduction degree are more dispersed. Compared with Under a microwave field, it is not the case that the more H 2 added to the CO atmosphere, the better the reduction effect. The atmosphere with the highest reduction was 50% H 2 + 50% CO (Figure 8b). The reason for this finding is most likely that CO changes the partial pressure of the atmosphere. CO and H 2 O are polar molecules, and polar molecules are susceptible to microwave radiation, which changes the balance of CO + H 2 O→H 2 + CO 2 , making the reduction reaction continue. Figures 9 and 10 show microstructural images of magnetite reduced by a mixed H 2 + CO atmosphere under a microwave field and conventional field, respectively. The edge of the Fe layer reduced by pure H 2 under a microwave field is thicker, and the Fe layer reduced by 60% H 2 + 40% CO is highly dispersed. Even if the particles are relatively small, the reduced Fe is highly dispersed. That is, the pores generated by the reduction of mineral powder particles with a high reduction degree are more dispersed. Compared with conventional heating, the bubble holes formed by reduced iron under a microwave field are larger under the same conditions.  Figures 9 and 10 show microstructural images of magnetite reduced by a mixed H2 + CO atmosphere under a microwave field and conventional field, respectively. The edge of the Fe layer reduced by pure H2 under a microwave field is thicker, and the Fe layer reduced by 60% H2 + 40% CO is highly dispersed. Even if the particles are relatively small, the reduced Fe is highly dispersed. That is, the pores generated by the reduction of mineral powder particles with a high reduction degree are more dispersed. Compared with conventional heating, the bubble holes formed by reduced iron under a microwave field are larger under the same conditions.  In the pure CO-reduced magnetite, the Fe phase in the microwave field forms blocks, while the Fe phase in the conventional field forms strips. When CO is added to the H 2 atmosphere under microwave heating, the reduction products swell. The more CO that is added, the more serious the degree of swelling. Additionally, 100% CO reduction ore powder will undergo severe swelling. Under the conventional field of CO reduction, swelling is not obvious. The swelling of magnetite during the reduction process is attributed to the tearing of the iron layer and the swelling of the ore layer caused by the increase in internal pressure in the void of the Wüstite/Fe phase boundary [32]. Swelling is the process of reducing product cracking. When the reduction degree is low, a high degree of crack swelling occurs. With the increase in the reduction degree, fragmentation enhances, and swelling weakens. The reason for the rapid swelling under microwave radiation is that the growth rate of the iron phase is fast. Microwaves increase swelling and cracking, and the addition of hydrogen weakens this behavior. When H 2 is added to CO, CO induces reductive swelling, swelling, and even stripping, while H 2 guarantees diffusion and reduction rates. Therefore, the reduction degree of H 2 and CO in the microwave field is higher than that of pure H 2 . Microwave treatment enhances the permeability of reaction gasses and circumvents the premature closure of surface pores under conventional thermal fields [33]. A microwave has the effect of accelerating reaction and diffusion, and when the reaction rate is low, it will accelerate diffusion sintering. Therefore, the iron phase topography is smoother, and the external morphology of the raw ore is no longer maintained.

Conclusions
(1) The effects of the H 2 and CO reduction of magnetite under a microwave field were compared. Under a microwave field, the metallization rate of the H 2 -reduced magnetite was 56% higher than that of CO at a high temperature. The metallization rate of magnetite reduced for 40 min by 60% H 2 -Ar at 1100 • C was 94%, and the reduction degree was close to 1. Under the same conditions, the metallization rate of CO reduction was 38%, and the reduction degree was 0.51.
(2) The microscopic morphologies of the particles' cross sections were observed. Under the condition of a 30% H 2 atmosphere, the increase in temperature was beneficial to the formation of a porous structure, and the coating layer was more easily formed at a low temperature (900 • C), while the small particle ore powder had almost no pores. For the reduction of ore powder by CO, the change in temperature and CO partial pressure produced neither coated layered iron nor pore structures. The formation of porous structures was affected by temperature and partial pressure. In addition, the effect of gas phase diffusion as a limiting factor of reduction is very critical. It is not easy to produce a porous structure when the degree of gas phase diffusion is limited.
(3) The microstructures of the particles were observed. The mineral particles after H 2 reduction had a porous honeycomb shape. The interconnected pore structure allowed the gas to continuously enter the reaction interface from the sample surface through the product layer. However, there were many iron whiskers on the surfaces of the minerals after CO reduction, and the whiskers and dense layer were not conducive to efficient reduction.
(4) The reduction of magnetite under a microwave field and a conventional field in a H 2 + CO mixed atmosphere was observed. Under conventional heating experimental conditions, as long as H 2 was added in a pure CO atmosphere, the higher the H 2 content, the better the reduction effect. Under a microwave field, the atmosphere with the highest reduction rate was 50% H 2 + 50% CO.

Conflicts of Interest:
On behalf of all authors, the corresponding author states that there is no conflict of interest to declare.